Crit Care Clin 23 (2007) 241–250

The Open Lung Concept of Mechanical

Ventilation: The Role of Recruitment
and Stabilization
Peter J. Papadakos, MD, FCCMa,b,c,*,
Burkhard Lachmann, MD, PhDd
a
Department of Anesthesiology, University of Rochester School of Medicine and Dentistry,
Box 604, 601 Elmwood Avenue, Rochester, NY 14642, USA
b
Department of Surgery, University of Rochester School of Medicine and Dentistry,
Rochester, NY 14642, USA
c
Department of Neurosurgery, University of Rochester School of Medicine and Dentistry,
Rochester, NY 14642, USA
d
Department of Anesthesiology, Erasmus-MC Faculty Box 2040, 3000 CA Rotterdam,
The Netherlands

Every year, worldwide millions of patients receive ventilator support in

the ICU. Mechanical ventilation remains the most common intervention
in the ICU and operating room to support patients in the acute and chronic
phase of care. A key measure of patient outcome and the quality of care pro-
vided by the ICU is ventilator-free days.
The use of ventilatory management protocols for ventilation of acutely ill
patients in the ICU has been evolving and improving continually. Strategies
have changed from optimizing convenient physiology variables, such as ox-
ygen and carbon dioxide levels, to protecting the lung from injury and
decreasing the cytokine modulation of the lung [1]. It has become clear,
however, that mechanical ventilation can attenuate lung damage and may
even be the primary factor in lung injury. This has led to renewed interest
in lung mechanics and ventilation. Much remains to be elucidated, however,
and ongoing debate continues.
One of the most important and recent areas of research and clinical interest
involves the strategy of lung recruitment. This strategy of lung recruitment or
open lung concept (OLC) refers to the dynamic process of opening previously
collapsed lung units by increasing transpulmonary pressure. The concept and

* Corresponding author. Division of Critical Care Medicine, University of Rochester,

Box 604, 601 Elmwood Avenue, Rochester, New York 14642.
E-mail address: peter_papadakos@urmc.rochester.edu (P.J. Papadakos).

0749-0704/07/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.ccc.2006.12.001 criticalcare.theclinics.com
242 PAPADAKOS & LACHMANN

its clinical use are not new. It first was proposed in the early 1990s [2]. Recent
experimental evidence only reinforces that this strategy may play an important
role in preventing ventilator-induced lung injury [3,4]. This article describes
the pathophysiologic basis and clinical role for lung recruitment maneuvers.
It reviews the literature and presents the authors’ clinical experience of over
15 years in the collaboration between Erasmus MC and the University of Ro-
chester. The authors are hopeful that these lung-protective strategies are pre-
sented in a useful format that may be useful to the practicing intensivist, thus
bringing laboratory and clinical research to bedside practice.

The physiology and pathophysiology of mechanical ventilation

Surfactant changes
It is known that the decrease in lung distensibility is a disturbed surfac-
tant system with elevated surface tension. This increase in surface tension
leads to an increase in forces acting at the air–liquid interface, resulting
finally in the end-expiratory collapse, atelectasis, an increase in the right-
to-left shunt, and a decrease in PaO2. This first was reported in the late
1950s, when Mead and Collier [5] reported that dogs that were ventilated
displayed a progressive decrease in pulmonary compliance and that these
changes were somehow related to the pulmonary surfactant system. Over
many years, two primary mechanisms of surfactant failure related to me-
chanical ventilation have been described.
In the first mechanism, mechanical ventilation enhances surfactant re-
lease from the pneumocyte type II into the alveolus [6]. This material sub-
sequently is lost into the small airways as a result of compression of the
surfactant film. The changes in alveolar surfactant may affect the permeabil-
ity of the alveolocapillary barrier to small solutes and proteins. The func-
tional integrity of both the endothelium and epithelium is a prerequisite
for maintaining a normal fluid balance at the alveolocapillary membrane;
therefore, these changes may cause the increased pulmonary leak observed
in respiratory failure and the formation of edema. Furthermore, surfactant
composition and function can be impaired by inhibitory factors from pro-
tein-rich pulmonary edema fluid or by the degradation in the alveolar space,
because of lipases and proteinases [7].
The second mechanism is based on the observation that the alveolar sur-
factant and the changes that are associated with mechanical ventilation may
result in the conversion of surface-active, large surfactant aggregates into
nonsurface-active aggregates [7,8]. Surfactant changes caused by mechanical
ventilation are reversible as a result of a metabolically active process involv-
ing de novo production of surfactant. The barrier function of surfactant
may collapse with barrier and mechanical ventilation, and there may be
transmigration of bacteria [9–11]. One of the key studies to develop this con-
cept of biotrauma was by Tremblay and colleagues [9]. Using isolated lungs,
 OPEN LUNG CONCEPT OF MECHANICAL VENTILATION 243

they investigated the effect of different ventilatory strategies on lung inflam-

matory mediator expression and production of tumor necrosis factor-a
(TNF a), interleukin-lb (IL-1b), IL-6, IL-10, platelet activating factor
(PAF), and interferon-gamma in the presence and absence of a pre-existing
inflammatory stimulus.
The use of high-peak inspiratory lung volumes and the avoidance of pos-
itive end-expiratory pressure (PEEP) during mechanical ventilation have
a synergetic effort on the release of proinflammatory mediators from the
lung tissue into the airways. But using 10 cm H2O of PEEP at comparable
peak inspiratory lung volumes or lowering peak inspiratory lung volume
when ventilating with zero PEEP reduced these cytokine levels. Therefore,
the lung is being identified as an important causative part of an inflamma-
tion-induced systemic disease state that can evolve to multiorgan failure
(MOF) rather than merely a pulmonary disease process. Alveolar collapse
with improper mechanical ventilation thus can lead to activation of a sys-
temic inflammatory response (SIRS) both locally and systemically, which
will play a role in modulating the individual patient’s outcome [12].

Modes of ventilation that will prevent ventilation-induced lung injury

Since its introduction in critical care medicine more that 45 years ago, ar-
tificial mechanical ventilation has been a standard lifesaving therapy. It was,
however, a topic of much discussion and controversy, because artificial ven-
tilation involves a disturbance of normal physiologic function of the lung.
Broad-based investigation into the most basic aspect of mechanical ventila-
tion tidal volume was published in the acute respiratory distress syndrome
(ARDS) network trial by the National Institutes of Health (NIH) only
6 years ago [13]. The study illustrated how the standard physiologic tidal
volume of 5 to 7 cc/kg had been adopted and the common practice of an
unnatural tidal volume of over 10 cc/kg. The authors are hopeful that this
natural tidal of 5 to 7 cc/kg has been accepted into practice and is common
practice in all ICUs. This simple change in practice will contribute greatly to
the outcome of ventilated patients.

Pressure-controlled ventilation
Pressure-controlled ventilation has been a mainstay in the treatment of
severe lung disease since described in the neonate by Colgan and colleagues
[14] in 1960 at the University of Rochester.
The keystone to proper mechanical ventilation is a pressure-controlled
platform. This mode of ventilation has many forms in modern ventilators
and differs in name by manufacturer. It is paramount that the user be facile
in the management of their ventilator fleet. Basic physiology serves as the
rationale for the use of pressure-controlled ventilation [15,16]. Because it
is established that artificial ventilation can both cause direct lung damage
244 PAPADAKOS & LACHMANN

and modulate cytokine release, it is imperative that one protects the lung from
ventilator-induced injury. Atelectasis not only affects local gas exchange but
also affects nonatelectatic areas [17]. The cycle of continuous expansion and
collapse of alveoli during the respiratory cycle creates a biologic stress. This
opening and closing affects structural changes by means of barotrauma and
volutrauma, as well as surfactant function and cytokine release.
If one evaluates mechanical ventilation in light of the basic Law of Lap-
lace (Fig. 1), one sees that using modes of ventilation that can control both
expiratory and inspiratory pressure may be an optimal way to ventilate the
lung. The rationale behind the high opening pressure to recruit the lung and
the need for lower pressures to keep the alveoli open can be deduced from
the pressure–volume curve of an individual alveolus. The Law of Laplace
links the pressure applied by the ventilator to alveolar pressure (P), which
relates surface tension (T) and radius (R):

P ¼ 2TR

In normal lung, alveolar surfactant plays a role in minimizing the surface

forces of the air–liquid interface, thus guaranteeing alveolar stability at all
alveolar sizes. In mechanically ventilated lungs, there may be varied levels
of dysfunction of the surfactant system, because of either direct ventilator
effects or the indirect effect of the systemic inflammatory response. The de-
gree of this surfactant dysfunction will determine the amount of pressure
needed to expand alveoli from closed to open.
Pressure-controlled ventilation allows the practitioner to control ventila-
tory pressure throughout the cycle, to generate the pressure necessary to

Fig. 1. Physiological behavior of the alveolus. The pressure–volume (P–V) relation on the X–Y
axes. The right side shows the status of the broncho alveolar unit. Its radius (r) reflects the P–V
relation (I-IV). Surface tension in pathological (T1) and normal conditions (T2) is shown. The
arrows indicate the direction from closed (bottom) to open (top) states and vice versa.
 OPEN LUNG CONCEPT OF MECHANICAL VENTILATION 245

expand alveoli. In true alveolar collapse, the pressure needed for alveolar re-
cruitment may reach levels above 70 cm H2O [3,4,18,19].
Alveolar beds may be opened best using a classic wave pattern of pressure
control, the decelerating wave pattern. This pattern is generated by pressure
differences between the inspiratory pressure delivered by the ventilator and
the pressure present inside the lung at the beginning of the inspiratory cycle,
resulting in minimization of flow. As the intrathoracic pressure increases,
the difference between the ventilator and the intrathoracic pressure dimin-
ishes, as does the resulting inspiratory flow. This decelerating flow pattern
is in contrast to the constant flow pattern that is used in most forms of vol-
ume-controlled ventilation.
A very important concept of pressure-controlled ventilation is fresh gas
distribution in the lung. The decelerating pattern opens alveoli better than
does a constant flow pattern. The resistance of airways influences the abso-
lute rate of the respiratory flow; therefore, if the resistance is high, the flow
will be reduced, and if the resistance is low, the flow will increase.
When new alveoli are recruited during an inspiratory cycle, the volume
necessary to fill these alveoli comes from the ventilator, which is the source
of the higher pressure, not from the adjacent lung units, because there is al-
ways equal pressure in all areas of the lung. Any reduction in alveoli size im-
mediately results in the flow of fresh gas from the highest pressure source,
which is always the ventilator into the alveolar unit. Decelerating wave pat-
tern pressure control ventilation also produces better pulmonary gas ex-
change through better gas distribution [15,20,21].
Volume control generates intrapulmonary redistribution of gas from
other hyperdistended lung units, the so-called Pendelluft effect. Pressure
control, in contrast, does not cause redistribution. Pressure-controlled
modes always generate an efficient system in which only fresh gas is entering
the recruited alveoli.

The open lung concept

Gattinoni and colleagues [4,22] showed that patients with early ARDS
had multiple areas of atelectasis, most commonly in the dependent lung re-
gions, which resulted in a reduced volume of aerated lung. These studies re-
vealed that the percentage of potentially recruitable lung varied widely
among patients with acute lung injury and ARDS, from a negligible fraction
to more than 50% of total lung weight. It is, therefore, recommended that
routine CT scanning be performed in all patients who have acute lung injury,
pulmonary contusion, chest trauma, and ARDS. This is very important, es-
pecially in managing complex patients who are failing therapy in the ICU.
The treatment for this alveolar collapse in clinical practice is lung recruit-
ment, the OLC first coined by Lachmann [2] in 1992. The clinician should be
aware that studies have shown that potentially recruitable lung is at a higher
percentage in patients with markedly poorer gas exchange and respiratory
246 PAPADAKOS & LACHMANN

mechanics, a greater severity of lung injury [4]. An association between the

percentage of potentially recruitable lung and the severity of lung injury, al-
though unexpected, appears logical. In healthy lungs, the percentage of po-
tentially recruitable lung is close to zero because of normally functioning
surfactant, which maintains alveolar units in a noncollapsed state. When
ARDS and other pulmonary pathologic states affect the lung, the extent
of inflammatory pulmonary edema is linked to the likelihood of gravity-de-
pendent alveolar collapse [23].
Thus, one must use the OLC to generate early lung recruitment and sta-
bilization, which brings the mechanically ventilated patient to near normal
physiology. The OLC of Lachmann slowly has found broad-based accep-
tance in the last few years [3,4,18,24]. The goal of OLC is simple: to have
as little or no atelectasis in the mechanically ventilated lung and also to gen-
erate as close to optimal gas exchange for each specific patient.
Intrapulmonary shunt ideally should be less than 10%, corresponding to
a PaO2 of more than 450 mm Hg, when breathing 100% oxygen at sea level [25].
In the past few years, OLC has led to the development of various tech-
niques to recruit and maintain a lung with minimal atelectasis [3,4,16,18,24].
The goal with all of these techniques is to minimize cyclic forces of alveolar
collapse and reopening. Despite the increasing body of literature on recruit-
ment, few studies have compared the various methods in terms of efficacy
and adverse effects. A common adverse affect of sustained high pressure
may be transient hypertension, but most recruitment techniques usually
call for only a few recruitment breaths. The authors have described using
only in the range of 10 breaths [16,24], which minimize this.
The choice of recruitment maneuver may depend on the individual patient
and the baseline ventilatory mode; a spontaneously breathing patient may not
tolerate a sustained high-pressure inflation, and a transient increase in PEEP
and peak pressure may be more appropriate in these patients. There is some
evidence that the type of lung injury (pulmonary versus extrapulmonary) may
affect tolerance to and efficacy of various recruitment modalities [26].
The frequency with which recruitment maneuvers must be applied is also
unknown. The authors’ ICU uses a recruitment maneuver when the patient ar-
rives to the unit and after any secretion management. Thus, this protocol must
be multidisciplinary with buy-in from physicians, nurses, and respiratory ther-
apy specialists. The authors have developed a simple way to understand the
concept. Fig. 2 shows the predetermined sequence of therapeutic phases,
each with its treatment objective. As shown in this figure, the goal of the initial
increase in inspiratory pressure is to recruit collapsed alveoli and to determine
the critical opening pressure. Then, the minimum pressure that prevents the
lung from collapsing is determined. Finally, after an active reopening maneu-
ver, sufficient pressure is implemented to keep the lung open.
After opening the lung and finding the lowest pressure to keep it open,
the resulting pressure amplitude is minimized and the same time pulmonary
gases improve after a successful alveolar recruitment. Should a renewed
 OPEN LUNG CONCEPT OF MECHANICAL VENTILATION 247

Fig. 2. Representation of the opening procedure for collapsed lungs. Note: the imperatives (!)
mark the treatment goal of each specific intervention. The bold words mark the achieved state of
the lung. At the beginning, the precise amount of collapsed lung tissue is not known.

collapse of alveoli occur, often caused by intrapulmonary suctioning or dis-

connection, a fall in PaO2 indicates that a reopening maneuver has to per-
formed in the same way as previously described. This is a highly important
point, in that secretion management must be balanced with alveolar recruitment.
Another important variable in the management of these patients is when,
in the process of the development of ARDS, the patient is received into care.
There are reports of higher response rates noted in patients early in the
course of their disease (eg, !72 hours) rather than later [27]. This probably
relates to the change in disease from an exudative to a fibroproliferative pro-
cess. Therefore, the authors believe that early application of OLC may not
only improve ventilatory mechanics but also affect outcome secondary to
cytokine modulation, bacterial translocation, and other factors [28,29].
OLC may be applied in at-risk patients during surgery [30,31]. It may be
an ideal technique in multitrauma patients with pulmonary injury as de-
scribed by Schreiter and colleagues [32] and others [33,34] who showed
that a recruitment maneuver applied at an early state of severe lung injury
can dramatically improve oxygenation recruit lung tissue and maintain
the newly recruited lung tissue. The authors, therefore, would include man-
datory recruitment procedures as a part of the treatment of suspected lung
contusion patients, thus obtaining and maintaining oxygenation values
above acute lung injury values, while minimizing further injury by applying
low tidal volume ventilation at minimal mean airway pressures. Guidelines
are available elsewhere [24].
Recruitment success and maintaining the OLC throughout the ventila-
tory status should be evaluated through ventilator graphics, CT scanning
and blood gas analysis. The management of patients with OLC is fluid,
and patients are evaluated and treated throughout their stay in the ICU.
248 PAPADAKOS & LACHMANN

General guidelines are simple to install in any ICU. The peak inspiratory
pressure (PIP) is adjusted to the lowest pressure, which keeps the lung open.
This lowest pressure is realized when the tidal volume remains stable, and
the arterial blood gases are constant. The ideal pressure is generally 15 to
30 cm H2O to prevent alveolar collapse. The level of PEEP should be ti-
trated through the use of best PEEP protocols to guide its use in conjunction
with recruitment protocols [35]. The advent of ventilator graphics has made
it much easier for respiratory care therapists to titrate to best PEEP to in-
dividual patient needs; this process is one of continuous re-evaluation.
Fig. 3 illustrates the pressure volume curve for the calculation of the inflec-
tion point.

Conclusion
The literature regarding the use of recruitment maneuvers is growing.
The pathophysiological rationale of cytokine modulation and compelling
laboratory and clinical trials support an open lung strategy in all mech-
anically ventilated patients in the ICU and operating theater. Through
real-time titration, the additional benefit of reduced ventilator-induced
lung injury can be accrued. It is essential to avoid doing harm by close mon-
itoring and ensuring that the overriding ventilatory strategy is one of pres-
sure limitation. Many questions remain, but the authors hope that the great
interest in the OLC over the last 5 years will generate an interest in the de-
velopment of clinical trials that will answer many of these questions in

Fig. 3. Best positive end-expiratory pressure calculation of inflection point.

 OPEN LUNG CONCEPT OF MECHANICAL VENTILATION 249

diverse patient populations, and also stimulate readers to develop a lung re-
cruitment protocol in their facilities.

Summary
The basic treatment principals therefore are:
Open up the whole lung with the required inspiratory pressures.
Keep the lung open with PEEP levels above the closing pressures.
Maintain optimal gas exchange at the smallest possible pressure ampli-
tudes to optimize carbon dioxide removal.
With the strict application of these principals, a prophylactic treatment is
available that is aimed at preventing ventilator-associated lung injury and
pulmonary complication without compromising optimal ventilation.

References
[1] Unilig S, Ranieri M, Slutsky AS. Biotrauma hypothesis of ventilator-induced lung injury.
Am J Respir Crit Care Med 2004;169:314–5.
[2] Lachmann B. Open up the lung and keep the lung open. Intensive Care Med 1992;18:81.
[3] Lapinsky SE, Mehta S. Bench-to-bedside review: recruitment and recruiting maneuvers. Crit
Care 2005;9:60–5.
[4] Gattinoni L, Caironi P, Cressoni M, et al. Lung recruitment in patients with acute respira-
tory distress syndrome. N Engl J Med 2006;354:1775–86.
[5] Mead J, Collier C. Relationship of volume history to respiratory mechanics in anesthetized
dogs. J Appl Physiol 1959;14:669–78.
[6] McClenahan JB, Urtrowski A. Effect of ventilation on surfactant and its turnover rate.
J Appl Physiol 1967;23:215–30.
[7] Haitsma JJ, Papadakos PJ, Lachmann B. Surfactant therapy for acute lung injury/acute re-
spiratory distress syndrome. Curr Opin Crit Care 2004;10:18–22.
[8] Verbrugge SJC, Bohm SH, Gommer D, et al. Surfactant impairment after mechanical ven-
tilation with large alveolar surface area changes and effects of positive end-expiratory pres-
sure. Br J Anaesth 1998;80:1–5.
[9] Tremblay L, Valenza F, Ribeiro SP, et al. Injurious ventilatory strategies increase cyto-
kines and C-fos MRNA expression in an isolated rat lung model. J Clin Invest 1997;99:
944–52.
[10] Van Bethmann AN, Brash F, Nusing R, et al. Hyperventilation induces release of cytokines
from perfused moose lung. Am J Respir Crit Care Med 1998;157:263–72.
[11] Kolobow T, Moretti MP, Fumagelli R, et al. Severe impairment in lung function induced by
high peak airway pressure during mechanical ventilation. An experimental study. Am Rev
Respir Dis 1987;135:312–5.
[12] Papadakos PJ. Cytokines, genes, and ARDS. Chest 2002;121:1391–2.
[13] The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as
compared with traditional tidal volumes for acute lung injury and the acute respiratory dis-
tress syndrome. N Engl J Med 2000;342:1301–8.
[14] Colgan FJ, Jorgensen SC, Lawrence RM. Maintenance of respiration in neonatal respiratory
distress syndrome. JAMA 1960;173:115–7.
[15] Papadakos PJ, Lachmann B, Bohm S. Pressure-controlled ventilation: review and new ho-
rizons. Clinical Pulmonary Medicine 1998;5:120–3.
250 PAPADAKOS & LACHMANN

[16] Papadakos PJ. Artificial ventilation. In: Hemmings H, Hopkins P, editors. Foundations of
anesthesia: basic and clinical sciences. London: Mosby; 2000. p. 507–14.
[17] Tsuchida S, Engelberts D, Peltekova V, et al. Atelectasis causes alveolar injury in nonatelec-
tatic lung regions. Am J Respir Crit Care Med 2006;174:279–89.
[18] Borges JB, Okmato GFJ, Matos M, et al. Reversibility of lung collapse and hypoxemia in
early respiratory distress syndrome. Am J Respir Crit Care Med 2006;174:268–78.
[19] Tremplay LN, Slutsky A. The role of pressure and volume in ventilator-induced lung injury.
Appl Cardiopulm Pathophysiol 1997;6:179–90.
[20] Munoz J, Guerrero JE, Escalante JL, et al. Pressure-controlled ventilation versus controlled
mechanical ventilation with decelerating inspiratory flow. Crit Care Med 1993;21:1143–8.
[21] Bohm S, Lachmann B. Pressure-control ventilation. Putting a mode into perspective. Inter-
national Journal of Intensive Care 1996;3:381–90.
[22] Gattinoni L, Pesenti A, Torresin A, et al. Adult respiratory distress syndrome profiles by
computed tomography. J Thorac Imaging 1986;1:25–30.
[23] Pelosi P, D’Andrea L, Vitale G, et al. Vertical gradient of regional lung inflation in adult re-
spiratory distress syndrome. Am J Respir Crit Care Med 1994;149:8–13.
[24] Papadakos PJ, Lachmann B. The open lung concept of alveolar recruitment can improve
outcome in respiratory failure and ARDS. Mt Sinai J Med 2002;69:73–7.
[25] Bohm SH, Vasquez de Anda GF, Lachmann B. The open lung concept. In: Vincent JL, ed-
itor. Yearbook of intensive care and emergency medicine. Berlin: Springer Verlag; 1998.
p. 430–40.
[26] Lim CM, Jung H, Koh Y, et al. Effect of alveolar recruitment maneuver in early acute respi-
ratory distress syndrome according to antiderecruitment strategy, etiological category of dif-
fuse lung injury and body position of the patient. Crit Care Med 2003;31:411–8.
[27] Grasso S, Mascia L, Del Turco M, et al. Effects of recruiting maneuvers in patients with acute
respiratory distress syndrome ventilated with protective ventilatory strategy. Anesthesiology
2002;96:795–802.
[28] Cakar N, Akinci O, Tugal S, et al. Recruitment maneuver: does it promote bacterial trans-
location? Crit Care Med 2002;30:2103–6.
[29] Chu EK, Whitehead T, Slutsky AS. Effects of cyclic opening and closing at low- and high-
volume ventilation on bronchoalveolar lavage cytokines. Crit Care Med 2004;32:168–74.
[30] Miranda DR, Gommers D, Papadakos PJ, et al. Mechanical ventilation affects pulmonary
inflammation in cardiac surgery patients: the role of the open-lung concept. J Cardiothorac
Vasc Anesth 2007; in press.
[31] Tusman G, Bohm SH, Vasquezde Anda GF, et al. Alveolar recruitment strategy improves
arterial oxygenation during general anesthesia. Br J Anaesth 1999;82:8–13.
[32] Schreiter D, Reske A, Scheibner L, et al. The open lung concept. Clinical application in tho-
racic trauma. Chirurg 2002;73:353–9.
[33] Ferguson ND, Chiche JD, Kacmarek RM, et al. Combining high-frequency oscillatory ven-
tilation and recruitment maneuvers in adults with early acute respiratory distress syndrome:
the treatment with oscillation and an open-lung strategy (tools) trial pilot study. Crit Care
Med 2005;33:479–86.
[34] Haitsma JJ, Bartels K, Papadakos PJ. Opening up to lung recruitment pulmonary contusion
and derecruitment: the role of inflammation. Critical Care and Shock 2006;9:9–15.
[35] Levy NM. PEEP in ARDSdhow much is enough? N Engl J Med 2004;351:389–91.